Scheduled Maintenance: We are aware of an issue with Google, AOL, and Yahoo services as email providers which are blocking new registrations. We are trying to fix the issue and we have several internal and external support tickets in process to resolve the issue. Please see: viewtopic.php?t=158230

 

 

 

Can Not-Calibrating a Battery Actually Save It?

Need help with peripherals or devices?
Post Reply
Message
Author
emariz
Posts: 2901
Joined: 2008-10-17 07:59

Can Not-Calibrating a Battery Actually Save It?

#1 Post by emariz »

Having used laptops as my primary systems for the last fifteen years, I have read a lot of recommendations (and myths) about batteries. Lately, I noticed that the battery's capacity might not have diminished as much as I thought, but it might simply not be properly calibrated.

This five-year-old, Sandy Bridge laptop (Vaio VPCEG13) still uses its original battery (a common 5,300 mAh / 11.1 V) and charger. And it works plugged in over 90 % of the time because it is used as desktop replacement.

Over the years, I have checked ACPI's statistics about the battery capacity and these days its (reported) capacity is 42 % of the original; which accounts for a bit more than two hours (it lasted around five hours when it was brand new.)

Code: Select all

$ acpi --details

Battery 0: Unknown, 99%
Battery 0: design capacity 4932 mAh, last full capacity 2082 mAh = 42%

Code: Select all

$ upower --dump

Device: /org/freedesktop/UPower/devices/battery_BAT0
  native-path:          BAT0
  vendor:               SONY Corp.
  model:                VGP-BPS26A
  serial:               7964
  power supply:         yes
  updated:              (101 seconds ago)
  has history:          yes
  has statistics:       yes
  battery
    present:             yes
    rechargeable:        yes
    state:               fully-charged
    warning-level:       none
    energy:              26,68 Wh
    energy-empty:        0 Wh
    energy-full:         26,71 Wh
    energy-full-design:  63,27 Wh
    energy-rate:         1,639 W
    voltage:             12,83 V
    percentage:          99%
    capacity:            42,2159%
    technology:          lithium-ion
For these reports of progressive capacity loss always seemed logical (even optimistic) along these years, I never doubted them and decided that as long as I got over 90 minutes of moderate use, I would not replace the battery. As I said, the laptop is almost always plugged in anyway.


Last weekend, I decided to test the power management action when the battery enters its critical low level, because the one thing worse than a dying battery is a system which does not save the current work before shutting down. To my surprise, the battery lasted over four hours. Yes, four!

The battery discharged properly and after two hours the system informed me that its capacity had reached 10 % and the charging LED started blinking, just as expected. But when the battery reached 8 %, it did not report further discharging, and I kept using the system for about two more hours. In the meantime, the system always showed the message 8% - under 10 minutes. Around the four-hour mark, it did reach 3 % and the system warned me that it would hibernate soon, a couple of minutes later (when it reached the 2 %), it entered (hybrid) hibernation.


The next morning, I retried this and got the same results. This time, I lowered the action-triggering percentages within UPower, so that the battery would actually drain completely and (hopefully) re-calibrate. It would inform me me at 5 %, warn me at 2 % and shut down at 1 %.

After two hours of use, the battery capacity was near its 10 % and the charging LED started blinking. Once again, the battery stopped informing further discharging under 8 %, and I kept using it two more hours. After about four hours of use, the system finally informed me (at 5 %), then warned me (at 2 %) and finally shut down (at 1 %). I could restart the system twice before the battery was finally depleted and the system shut down automatically each time.
I let the battery rest for over eight hours before recharging it.


Today, I hoped that the battery would be recalibrated, but, as you can see in the output of ACPI and UPower, it is not. It still ran for four hours of use though, yet the battery would stop informing further discharging under 8 %, until it did reach its critical percentage.


Some thoughts:

1. Using the laptop plugged in over 90 % of the time is the number one suspect of messing with the battery calibration.
2. Being almost always plugged in may also translate into very few discharge cycles.

3. For the progressive loss of the battery capacity seemed logical, I always charged it when its percentage was (reported as) low (i.e. after 2:30 hours or so).
4. If the battery has not been properly calibrated for a long time, I might have been re-charging it at a high percentage of its actual capacity (say 40 %, or after 2:30 h of its 4 h duration.)

5. For re-charging at high percentages is often recommended for preserving the battery lifespan, I might have unknowingly saved it.

User avatar
Head_on_a_Stick
Posts: 14114
Joined: 2014-06-01 17:46
Location: London, England
Has thanked: 81 times
Been thanked: 133 times

Re: Can Not-Calibrating a Battery Actually Save It?

#2 Post by Head_on_a_Stick »

Fascinating post, thanks for this.

My Haswell laptop has also been plugged in almost continuously and I have now taken to running it without a battery because the life (according to the percentage readout gleaned from /proc) had gone from 4hrs to ~2.5hrs.

I will have to try a full discharge tonight.

I have noticed that systemd will cleanly shut the system down automatically in the event of complete battery discharge, which is nice :)
deadbang

emariz
Posts: 2901
Joined: 2008-10-17 07:59

Re: Can Not-Calibrating a Battery Actually Save It?

#3 Post by emariz »

Head_on_a_Stick wrote:My Haswell laptop has also been plugged in almost continuously and I have now taken to running it without a battery because the life (according to the percentage readout gleaned from /proc) had gone from 4hrs to ~2.5hrs.
Hopefully, the battery will not be calibrated and reports less capacity than it actually possesses.
The durability of this Vaio has been surprising, up there with old IBM ThinkPads. I upgraded its RAM and replaced the HD with an SSD, and it hides its age pretty well.
Head_on_a_Stick wrote:I have noticed that systemd will cleanly shut the system down automatically in the event of complete battery discharge, which is nice.
Why does it not enter hibernation or hybrid-hibernation? I cannot see the point of its closing everything and shutting down. One only witnesses a critical-level action precisely when one is using it; and I would like to continue working exactly where I was.

User avatar
stevepusser
Posts: 12930
Joined: 2009-10-06 05:53
Has thanked: 41 times
Been thanked: 71 times

Re: Can Not-Calibrating a Battery Actually Save It?

#4 Post by stevepusser »

Isn't the system getting its battery charge % from the battery voltage, assuming it's nearly completely discharged when it nears the lowest limit? Maybe you have one cell that's going bad, and the others are still good, leading to an overall quick drop in voltage.
MX Linux packager and developer

User avatar
Head_on_a_Stick
Posts: 14114
Joined: 2014-06-01 17:46
Location: London, England
Has thanked: 81 times
Been thanked: 133 times

Re: Can Not-Calibrating a Battery Actually Save It?

#5 Post by Head_on_a_Stick »

Head_on_a_Stick wrote:I will have to try a full discharge tonight.
The battery status was accurately reported all the way down to 2%, at this point the system suspended automatically; the battery life was 3 hours (4 hours when new).

CLEVO W550SU
emariz wrote:Why does it not enter hibernation or hybrid-hibernation?
The last time I tried that I had no swap or swapfile and I had disabled suspend.

My primary concern is filesystem damage, a random shutdown is not a problem for my "work"flow :)
deadbang

User avatar
bw123
Posts: 4015
Joined: 2011-05-09 06:02
Has thanked: 1 time
Been thanked: 28 times

Re: Can Not-Calibrating a Battery Actually Save It?

#6 Post by bw123 »

Ok, here's my weird battery story, acpi -i reports 41% charged but when I run on battery, the battery monitor starts at 100% and runs down until about 55% then drops to 5% so I'm thinking it really means (current chg-41%) the whole time.

I just went ahead and bought a new one for 30 bucks.
resigned by AI ChatGPT

Post Reply